80 LEGUMES'OS.E. 



racemis axillaribus 2-4-fasciculatis pedicellis 1 cm. longis floribns 2-S- 

 fasciculatis, bracteis obsoletis, petalis ad marginem purpurascentibus. Calyx 

 breve cylindricus, apice sabtnmcatus, 5-dentatus, dentibus posterioribtis 

 subobsoletis, 3-anterioribus oonspicuis, inedio pins conspicuo. Yexillum 

 obovoideum apice emarginahmi ; alre leviter longiores inedio cum carinis 

 tenuiter conniventes ; carime apice liberse, basi distinctce. Stamen vexillo 

 oppositum liberum, cfetra coimata. Stylus filiformis postice reeurvus, ant ice 

 deorsum ciliolatus ; stigma terminate, punctiforme ; ovarium sessile, ovnlis 5. 

 Frnctus oblongi subcomplanati obliqni 4 cm. tougi 3 cm. lati 23mm. 

 crassi 1-spermi exalati lignosi extns dense tuberculati, prope suturam 

 sublgeves et obscure costati, in maturo rarius ad suturam inferiorem 

 deliiscentes, apice rotundati ad stimmum breve apiculati, basi obtusi vel 

 breviter stipitiformes angustati mcdio sensim et obscure tenuissime vix vel 

 hand constricti, pedicellis 1 cm. longis validis. Semina obreniformia 3 cm. 

 longa 2-i cm. lata ; testa nitida nigricans, ad apicem medic > et ad centrum pro- 

 funde impressa; cotyledonibus amplis late oblongis 3 cm. latis 22mm. longis 

 basi profunde cordatis, sinibus 14 mm. longis ; radicula obsoleta. 



HAB. Sankakuyu, leg. K. NAGAI, 1902. 



The present plant is one of those leguminous plants which afford a 

 poisonous matter used for fishing by the aborigines in Formosa, It was 

 first labelled by Prof. J. MATSUMURA as Derris chinensis BENTH. Soon after, 

 the professor regarded the plant different from BENTHAM'S species and 

 named it D. tcdivaniana. The same plant was studied by Dr. K. NAGAI 

 who regarded it to be referable to LoncJiocarpus or more probably to 

 Coiibulandia, but not to Derri. He then expressed this opinion in a 

 number* of the Journal of the Tokyo Chemical Society. In my opinion, 

 the plant certainly does not belong to Derris on account of its wingless 

 fruits ; nor should it be referred to LoncJiocarpus, for the plant has a sessile 

 ovary, but has neither elongated pod nor broaded suture. It is also different 

 from Coubulandia in having neither torulose nor rounded pods. A genus to 

 which the present plant should probably be referred is, in my opinion, 

 Pongamia, as is seen from the structure of the fruits. The descriptions of 



* Journal of the Tokyo Chemical Society XXIII 7, pp. 744-777 (Japanese). 



